deviation of the Gaussian Kernel to be used to regularize the gradient and the radius of the window to be used for evaluating the local normalized cross correlation. """ sigma_diff = 3.0 radius = 4 metric = CCMetric(2, sigma_diff, radius) """ Let's use a scale space of 3 levels """ level_iters = [100, 50, 25] sdr = SymmetricDiffeomorphicRegistration(metric, level_iters) sdr.callback = callback_CC """ And execute the optimization """ mapping = sdr.optimize(static, moving) warped = mapping.transform(moving) """ We can see the effect of the warping by switching between the images before and after registration """ regtools.overlay_images(static, moving, 'Static', 'Overlay', 'Moving',
deviation of the Gaussian Kernel to be used to regularize the gradient and the radius of the window to be used for evaluating the local normalized cross correlation. """ sigma_diff = 3.0 radius = 4 metric = CCMetric(2, sigma_diff, radius) """ Let's use a scale space of 3 levels """ level_iters = [100, 50, 25] sdr = SymmetricDiffeomorphicRegistration(metric, level_iters) sdr.callback = callback_CC """ And execute the optimization """ mapping = sdr.optimize(static, moving) warped = mapping.transform(moving) """ We can see the effect of the warping by switching between the images before and after registration """ regtools.overlay_images(static, moving, 'Static', 'Overlay', 'Moving',
